sign in sign up
<<
Home , Beaver, North American beaver, Slough Creek (Wyoming)

The History and Current Status and Distribution of in Yellowstone National Park Author(s): Douglas W. Smith, Daniel B. Tyers Source: Northwest Science, 86(4):276-288. Published By: Northwest Scientific Association DOI: http://dx.doi.org/10.3955/046.086.0404 URL: http://www.bioone.org/doi/full/10.3955/046.086.0404

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. Douglas W. Smith1, National Park Service, Yellowstone National Park, PO Box 168, Yellowstone National Park, 82190 Daniel B. Tyers, U.S. Forest Service, Gallatin National Forest, Bozeman, 59715

The History and Current Status and Distribution of Beavers in Yellowstone National Park

Abstract Despite Yellowstone National Park’s (YNP) long history and well studied large and vegetation, beavers (Castor canadensis), an important ecosystem driver, have received relatively little study. We summarize population surveys of beavers that began in 1921 and continued up to the present. The first surveys (1921 and 1923) were from the ground and conducted in a limited area in the northern portion (northern range; NR) of the park. Twenty-five colonies were found and beavers were considered abundant and using ( tremuloides) and (Salix spp.) as a food source and building material. A follow up survey in 1953 found 6 NR sites, but none of the earlier sites from the 1920s were active and no aspen use was reported. Some locations were reported from the park interior. A limited ground survey was conducted in 1979-80. In 1988-89 and 1994 two incomplete, mostly ground surveys were conducted and estimated 71 and 44 colonies, respectively, in YNP. In 1996, 1998, 1999, 2001, 2003, 2005, 2007 and 2009 complete, park-wide aerial surveys were conducted and active colonies ranged from 44 (1996) to 127 (2007) with an increasing trend. Therefore, in a period of about 90 years (1920s–2000s) the population in the northern portion of the park appears to have declined then increased probably because of a willow recovery.

Keywords: beavers, Castor Canadensis, Yellowstone, willow, aspen

Introduction wide trends suggesting a reduction of beavers in the mid- to late 1800s (Schullery and Whittlesey Once widely distributed across , 1992). After this time, the first survey of beavers beavers (Castor canadensis) were much reduced conducted in the park was in 1921 (Warren 1926) during Euro-American settlement of the continent which found abundant beavers in the Tower Junc- (Baker and Hill 2003). Some have argued that tion area (a region in northern YNP). Starting with beaver , used to make felt for hats in Europe, this first survey in 1921, our goal was to summarize was the major impetus for exploration of the New subsequent efforts searching for beaver colonies World which resulted in European colonization (and sign of them) across YNP. These searches (Innis 1930, Nute 1978). Consequently, beavers were recorded in unpublished park reports and were extirpated through human and hunt- a thesis, which we combined with more recent, ing from vast regions of North America and often rigorous surveys in order to create a population there were no or poor records (typically journals history of beavers in YNP. only) of their abundance and distribution (Schul- Summarizing beaver population records in lery and Whittlesey 1992). Despite Yellowstone Yellowstone is important because there are many National Park’s (YNP) early establishment in 1872, disparate notes and reports creating a confusing essentially during the western , records history (Table 1). Ernest Thompson Seton first of beavers in YNP are similarly opaque coming noted beavers in the late 1890s in the Tower Junc- primarily from journals of trappers and early park tion area (cited in Warren 1926). This attention at administrators (Schullery and Whittlesey 1992). Tower Junction probably caused Edward Warren Park records, however, are reflective of region- (1926) to survey beavers in 1921 and 1923, partially because some worried they would cut down all the 1Author to whom correspondence should be addressed: aspen (Warren 1926). In the of 1930–1931, Email: [email protected] E. Roy Arnold (1931) unsystematically checked the

276 Northwest Science, Vol. 86, No. 4, 2012 TABLE 1. Surveys for beavers conducted in Yellowstone National Park 1921-2009.

Investigator Years Surveyed Method Area SurveyedA Citation E. Warren 1921 and 1923 Ground Tower/Mammoth Warren 1926 E. Arnold 1931 Ground Tower Area Arnold 1931 R. Jonas 1953-1954 Ground Tower/Mammoth Jonas 1955 Some Interior S. Fullerton 1979-1980 Ground Park-wide Fullerton 1980 Consolo Murphy 1988-1989 Ground Park-wide Consolo Murphy and Hanson 1993 Consolo Murphy 1994 Ground Park-wide Consolo Murphy and Tatum 1995 Smith 1996, 1998, 1999, 2001, Aerial Park-wide This Study 2003, 2005, 2007, 2009

A None of the surveys preceding the aerial surveys conducted by Smith were considered complete and only represent partial coverage of Yellowstone National Park. The Smith 2005 survey also was not complete due to early ice causing cessation of survey effort. status of some of the Warren colonies around Tower Besides summarizing the beaver population Junction. In 1953 Robert Jonas, a graduate student history in YNP, we discuss willow (Salix spp.) at Montana State University (MSU), resurveyed and aspen (Populus tremuloides) abundance and the areas reported on by Warren (Jonas 1955) with distribution because it is the primary winter food the objective to evaluate the status of beavers and of beavers and a widely published on topic that condition of the habitat since the Warren surveys. is intensely debated (Singer et al. 2000, Wagner Jonas expanded his survey area slightly, mostly 2006, et al. 2007, Kauffman et al. 2010). It through interviews with park rangers, but he did is debated because declines in willow and aspen not systematically survey the northern portion of over the past 100 years are attributed to various the park (an area now referred to as the ‘northern causes (National Research Council 2002). Some range’ (NR) and the wintering area for many have argued that the NR ecosystem of YNP at the Yellowstone ungulates; Houston 1982, Lemke beginning of the 20th century was in a vastly dif- et al. 1997) or the park interior. All of these were ferent condition, an ecological state dominated by ground surveys—the observers walked (or rode willow, aspen, and beavers with some elk (Cervus horseback) the area of interest. Another ground elaphus) and more carnivores (e.g., , Canis survey occurred in 1979–1980 that again focused lupus). In the last 100 years carnivore reductions, on the NR area of Warren, Arnold, and Jonas, but which allowed ungulates to increase, transitioned this too was incomplete, lacked a finished report the to a shrub–steppe–grassland domi- (Fullerton 1980), and often referred to beaver sign nated by elk (Wolf et al. 2007, Wagner 2006). and not colonies. These authors suggest that beavers may have been In 1988–1989 and 1994 National Park Service critical to maintaining this willow–aspen ‘state’ (NPS) biologist Sue Consolo Murphy expanded through a positive feedback cycle involving beaver the search area for beavers and looked across the hydrology. The loss of beavers may have park, mostly from the ground like Warren and contributed to altering this system. Jonas, but was the first to use aircraft (Consolo Others have argued that willow and aspen have Murphy and Hanson 1993, Consolo Murphy always been a minor component of the vegeta- and Tatum 1995). From 1996–2009, the senior tion on the NR, never occupying more than 10% author (D.W. Smith) conducted intensive aerial of the entire area (Houston 1982, Despain et al. surveys of the entire park searching every other 1986, YNP 1997). This argument further states, year with the objective of a complete count and and despite declines of 50–100% in area occupied to detect population trend (Table 1, Smith and by willow/aspen in the last 100 years (Wagner Tyers 2008). 2006), that little of this vegetative reduction can

Beaver Abundance and Distribution in Yellowstone 277 be attributed to loss of beavers and increases in 3) review the willow and aspen literature from elk but is mainly due to fire suppression and cli- the NR of YNP to aid interpretation of beaver mate (Houston 1982, Despain et al. 1986). This population fluctuations. was one of the most argued of all YNP issues (Houston 1982, Singer et al. 2000, Despain et Study Area al. 1986, Kay 1990, YNP 1997, Wagner 2006 to The NR and interior of YNP have been described mention only some). Although usually mentioned, in numerous publications (e.g., Houston 1982, De- discussion concerning beavers is underrepre- spain 1990, Smith and Siegel 2000), but in general sented compared to ungulates, arguably almost these two systems can be contrasted based on eleva- neglected and certainly their role in vegetation tion, vegetation, and geology (Smith et al. 2004). changes, or response to them. This is especially The NR is lower elevation (1500–2200 m) with interesting now because willow for the first time primarily shrub–steppe, Douglas-Fir (Pseudotsuga in decades has begun growing taller (Ripple and menziesii), and lodgepole (Pinus contorta) Beschta 2004a, Beyer et al. 2007), allowing a at higher elevations. Riparian areas, highly linear resurgence in the beaver population on YNP’s NR. in , have willow, which have declined and This willow release is coincident with mid-1990s been height limited for several decades until the wolf reintroduction (Bangs and Fritts 1996), and late 1990s release (Beyer et al 2007, Wolf et al. there is some evidence that approximately 10 2007, Bilyeu et al. 2008). Aspen, which occurs years after willow (mid to late 2000s), aspen may over < 2% of the area, and cottonwood (Populus also be releasing (Ripple and Beschta 2007; E. angustifolia and P. trichocarpa) have both de- Larsen, University of Wisconsin-Stevens Point, clined since the Warren (1926) and Jonas (1955) personal communication). This willow/aspen studies (Houston 1982, Kay 1990, Beschta 2005, resurgence has sparked a new debate as to the Wagner 2006). Little to no recruitment has been cause of the release, one that extends historical reported for several decades (excepting post-1988 arguments centering on elk, climate, and fire (or fire affects; Larsen and Ripple 2003, Kauffmann a combination of these) to now invoking wolves et al. 2010). Like willow, however, but about a (Beyer et al. 2007, Ripple and Beschta 2004b). decade later, a release of aspen stems may be oc- We seek to place beavers in this debate as they curring in some areas (Ripple and Larsen 2000, may be important to any argument. Ripple and Beschta 2007, E. Larsen, University An already complicated story is made more of Wisconsin-Stevens Point, personal communica- so because beavers were reintroduced upstream tion). Contrasted with the NR, the YNP interior of YNP by Gallatin National Forest (GNF) from is a high plateau fringed by mountains on the 1986–1999. This reintroduction provided the east and northwest (2200–3000 m) with mostly colonizers that contributed to the YNP popula- coniferous vegetation (primarily lodgepole pine tion increase, but this increase could not have but with some Engelmann [Picea engelman- occurred had the habitat not recovered first (Smith nii] and subalpine–fir [Abies lasiocarpa] and high and Tyers 2008). Our focus is YNP, but GNF’s elevation whitebark pine [Pinus albicaulis]) with reintroduction of beavers played a significant some exceptions in the southeast, southwest, and role in the recolonization of YNP’s NR, so beaver northwestern portions of the park where willow is re-occupation of GNF needs to be included in the found. Extensive stands of willow exist in areas YNP NR account. along the south of Yellowstone Therefore, given this complicated debate and , especially on the as it flows into unreported history, our objectives were to: 1) Yellowstone Lake, and other stands occur but determine beaver abundance, distribution and less extensively in the Thorofare, Bechler, and population trend in YNP from 1996–2009; 2) drainages in the northwest region of the park. summarize early beaver censuses and compare There is virtually no aspen available to beavers in to contemporary ones to reconstruct beaver abun- this interior region. Aspen is present but uncom- dance and areas of use over the last 90 years; and mon and out of reach of beavers due to previous

278 Smith and Tyers cutting activity. This is especially evident along ferent locations, but when discussing the beaver northwestern drainages (e.g., , Campanula, population (Table 2) report on the total for YNP. Gneiss creeks). Warren (1926) and Jonas (1955) used ground In general, gradients across the park ex- surveys on the NR and did not survey the entire ceed the slope that beavers can occupy (Baker and area. Some of Jonas’ locations were from the park Hill 2003, Hay 2010). Some exceptions occur, for interior and were identified through interviews with example along the Yellowstone River throughout park staff. Fullerton (1980) again used ground sur- its course across the park, (excepting the Grand veys and searched for areas of beaver activity and Canyon of the Yellowstone) and other rivers like did not attempt to assess active colonies, although the Gallatin, Lamar, Snake, Firehole, Madison, they were mentioned. Both Consolo Murphy led and Gibbon. Some creeks and , especially surveys searched from the ground across the park, Slough Creek, also have habitable gradients for and developed a park staff reporting system to beaver settlement. There are extensive low gradient locate areas of beaver activity and report sight- areas along these waterways, however, that do not ings. Each of her surveys augmented the ground have vegetation suitable for beavers, so the beaver effort with one fixed-wing flight which was not habitat requirement of low gradient combined systematic but checked areas surveyed from the with suitable woody vegetation (or aquatics) is ground. It was explicitly stated that neither the limited across most of Yellowstone. 1988–89 (Consolo Murphy and Hanson 1993) or 1994 surveys (Consolo Murphy and Tatum 1993) Methods estimated the beaver population or trend. None of the above surveys covered the entire park and Previous Beaver Surveys most areas were searched in summer to early fall We reviewed the published, but not peer reviewed, and it is well known that mid- to late-fall is a more reports of Warren (1926), Jonas (1955), and appropriate time to count beaver colonies (Hay Consolo Murphy and Hanson (1993) and the 1958, Baker and Hill 2003). unpublished reports of Arnold (1931), Fullerton Definitions of a beaver colony were not con- (1980), and Consolo Murphy and Tatum (1995) sistent between the observers, and not always for descriptions and locations of beaver colonies clearly reported. Rather, beaver sign or activity throughout YNP. These reports represent six in- was discussed sometimes without concluding that complete surveys for beavers across YNP from a beaver colony was present, or the activity was 1921–1994 (Table 1). We mapped each survey found in summer which is not always indicative of beavers using an active colony in autumn as a of colony presence (Smith 1997, Baker and Hill unit of measure with ArcMap 9.3 (Environmental 2003). The colony definition we used was a fresh Systems Research Institute, Redlands, CA). An lodge and/or recent beaver sign with a food cache active colony was sometimes not clearly defined in October (Shelton 1966, Hodgdon and Lancia by the various authors, many just reported beaver 1983, Jenkins and Busher 1979). Fresh lodges are sign, so some interpretation was necessary (see easy to spot from aircraft because recent repair definition below). We also noted what types of by beavers adds sticks (freshly peeled ones shine woody vegetation was being utilized by beavers, bright white) and mud to the structure, while inac- as it was often mentioned in the reports. tive lodges have vegetation growing from them We divided the park into two units: the northern with old, faded pieces of wood. Occasionally a range (NR) and the interior. We did this because the lone beaver or pair will renovate an old lodge, but areas are often considered as two separate systems some signs of use will usually be visible. Critical (Houston 1982, National Research Council 2002) to the assignment of activity is the presence of a based upon differences in elevation, vegetation, food cache in fall as this almost always indicates geology, and wildlife (Despain 1990, Smith and presence of a colony (Jenkins and Busher 1979, Siegel 2000). Throughout we refer to these dif- Hodgdon and Lancia 1983, Baker and Hill 2003).

Beaver Abundance and Distribution in Yellowstone 279 TABLE 2. Location of beaver colonies in Yellowstone National Park, 1996-2009.

______Number of Active Colonies______Location 1996 1998 1999 2001 2003 2005 2007 2009 Northern Range Glenn Creek 0 0 0 0 1 * 1 0 Slide Lake 1 1 0 0 0 * 0 0 Gardner River 0 0 0 0 0 0 0 0 Yellowstone River 0 0 0 0 0 1 0 0 Elk Creek 0 0 0 0 0 0 1 0 Lamar Valley 0 0 0 1 1 * 3 1 Slough Creek 0 0 1 3 6 9 6 8 Crystal Creek 0 0 0 0 0 0 0 1 Soda Butte 0 0 0 0 0 0 0 1 Round Prairie 0 0 0 0 0 0 0 1 Northwest Bacon Rind Creek 0 0 0 0 1 1 3 0 Fan Creek 0 0 0 0 0 2 5 4 Gallatin River 2 2 5 6 1 7 7 9 Grayling Creek 3 0 1 6 5 5 5 4 Panther Creek 0 0 0 0 0 0 0 1 Willow Park * 2 4 3 4 4 6 4 Campanula/Gneiss/ Creek 7 6 7 8 10 15 16 13 Creek 4 7 11 9 3 4 5 6 Maple Creek1 4 6 7 6 Harlequin Lake 0 1 1 1 1 1 1 1 Madison River 3 3 3 3 1 2 1 0 Southwest Bechler River 1 0 3 2 3 3 1 2 Boundary Creek 2 2 1 3 7 0 6 2 Falls River 0 2 3 6 3 2 6 2 Mountain Ash/Proposition Creek 7 6 6 1 0 3 1 1 Other Bechler 0 1 1 0 1 0 2 2 Southeast/Southcentral Hayden Valley 0 0 0 1 0 * 0 0 Creek * * * * 0 0 2 0 Basin Creek * * * * 0 0 3 3 Grouse Creek 1 0 0 0 1 * 1 0 Outlet Creek 0 1 0 0 0 * 1 1 Heart Lake Area (Surprise/Beaver/Aster Cks) 0 0 0 0 6 * 2 4 Shoshone (/Delacy) 0 1 1 1 0 * 2 2 South Lewis Lake Area 0 0 0 2 2 * 0 0 Snake River 3 2 4 4 3 * 3 0 Yellowstone River Area (Delta) 15 14 23 17 21 * 29 36 Chipmunk Creek 0 0 2 0 0 * 1 1 Total 49 51 77 77 85 65 127 116

1For years 1996-2001, Maple Creek data was tabulated under Campanula/Gneiss/Duck Creek or Cougar Creek. *Not censused

280 Smith and Tyers Often beavers in YNP use bank dens without con- Results structing a lodge, so a fresh food cache along the bank with sign was the definitive measure of an Previous Beaver Surveys active colony. This definition was applied when In 1921 Warren (1926), by our definition, found interpreting all of the previous literature. 25 colonies on the northern range and a partial survey in 1923 of the same area found 9 colonies New Beaver Surveys (Figure 1). He made no attempt at a complete In 1996, 1998, 1999, 2001, 2003, 2005, 2007, survey of the NR, but rather recorded activity at and 2009 we counted beaver colonies across the colony sites that were accessible from the Tower entire park using fixed-wing aircraft (Supercub Junction/Mammoth areas (Figure 1). He did not PA-18) following standard methods published survey the park interior. He noted that most of the for aerial beaver surveys (Hay 1958, Swenson et colonies were cutting aspen trees. al. 1983). Our objective was a complete count of In 1931, park ranger Arnold stationed at Tower YNP. Most of the time an active beaver colony was Junction and near where Warren had conducted his easily identified by the types of sign mentioned surveys, opportunistically surveyed some of the in the previous section. In < 5% of the cases > 1 Warren areas (specifically the areas from Geode food cache was recorded and attributed to just one Creek east to the Lamar–Yellowstone confluence colony. Beaver territoriality probably prohibits between the park road and the Yellowstone River). two colonies being closely located within 50 m He mostly reported beaver activity and sign, and (Jenkins and Busher 1979, Hodgdon and Lancia mentioned ‘many active colonies’, however neither 1983, Baker and Hill 2003), and often two food his notes nor map allow a count of how many caches were observed within this distance. In these colonies were present. He also mentioned much cases we attributed the two caches to one colony old beaver activity in these areas and a ‘depleted’ (Baker and Hill 2003). While flying over each food supply at one colony site near Geode Creek colony site, we marked the location with a Garmin and only available at Elk Creek. At another GPS unit that was later downloaded into ArcMap site (Crescent Hill) he mentioned ample aspen. so colony locations could be plotted on a map. The aerial surveys were always conducted in In 1953, Jonas (1955) repeated the Warren’s late fall (typically no earlier than mid–October) survey. In addition to revisiting the Warren sites, after the and had shed their leaves he expanded the survey to other parts of the park (which enhanced sightability) and the construction including some areas from the park interior. He of food caches were well underway and easily found no beavers at the Warren colony sites on observed from aircraft (Baker and Hill 2003, the NR. He found 8 other northern range sites Smith 1997, Hay 2010). In 2005 early onset of that were not reported by Warren (Figure 1). He ice curtailed the survey so the entire park was not reported little beaver use of aspen and reported counted that year. Each survey took multiple flights on the decline of aspen and beavers on the NR. of 3–4 hours each and survey time ranged from He reported 13 other sites from the park interior 12–16 hours/survey. The same pilot and one of (Figure 1). four observers flew each survey, but one observer In 1979–1980 Fullerton (1980) ground- (D.W. Smith) flew > 80% of the survey routes. searched for beaver colonies in the Warren/Jonas Surveys focused on areas of appropriate habitat sample area and found only one occupied site at and areas of historical beaver occupation. More Lost Lake. Jonas did not record this site as active. time was spent flying waterways the first two In the Mammoth area she recorded beaver activity surveys to ensure that some unlikely areas did in the Gardner River, at the Beaver , Crevice not support beavers, later these areas were not Lake, and Willow Park. Several areas of beaver surveyed. We also did not survey streams of high activity were recorded along the , gradient (> 6%; Hay 2010) and/or sites with no especially in the area around the confluence with deciduous or aquatic vegetation. (but none along the creek itself).

Beaver Abundance and Distribution in Yellowstone 281 Figure 1. Distribution of beaver colonies in Yellowstone National Park 1921-2009. Surveys are not comparable because some did not cover the entire park or used different methods (i.e., censused via ground vs. with aircraft (see notes under each map). The 2005 survey was complete except for southeast region of the park, so colony count is not actual total, but counts and distribution of other areas is comparable to other years. Definition of a beaver colony was standardized for this paper (presence of a food cache) as not all previous workers had a consistent definition. Shaded region represents the area referred to in the text as the northern range.

Notably, reports on beaver occupation in Slough to identify these areas as dense centers of beaver Creek were inconclusive, but it does not appear activity and occupation. Other areas of beaver that she surveyed the upper (northern) reaches of activity were Pelican Creek, Hayden Valley (Yel- the creek. She recorded aspen use at only one site lowstone River), Canyon area (Cascade, (Beaver Ponds) otherwise beavers were feeding on Grebe, Cygnet), and Harlequin Lake. She noted willow (or aquatics, see below). In the interior she use of aquatics at several sites (Obsidian, Cascade, found numerous areas of beaver use especially in Grebe, Beaver, Harlequin lakes) the southeast, southwest, and northwest regions Consolo Murphy and colleagues conducted the (as far south as the Madison River where extensive most extensive ground survey for beavers to date activity was noted), and this was the first survey in 1988 and 1989 (Consolo Murphy and Hanson

282 Smith and Tyers Figure 2. Distribution of beaver colonies and willow stands in Yellowstone National Park in 1996 (Smith and Tyers 2008).

1993) and again in 1994 (Consolo Murphy and 2). No colonies were found at the Warren/Jonas Tatum 1995). They found 71 beaver sites (colony sites, except possibly one along lower Elk Creek or beaver sign) park-wide in 1988–89 (Figure 1) which was a site discussed by Warren and Jonas, which included a complete count of the NR in the but it was not clearly stated whether or not a colony Warren/Jonas areas, but they did not specifically existed there in the 1920s and 1950s (Warren 1926, mention the colonies surveyed by them, and they Jonas 1955; Figure 1). Warren (1926) mentioned appeared to be inactive from their figure (their that Ernest Thompson Seton recorded a colony on Figure 2). They repeated the survey in 1994 Elk Creek in the late 1800s, and it may have been (Consolo Murphy and Tatum 1995) and found in the same area that was found to have beavers 44 colony/beaver sites with no occupation in the intermittently in the 2000s. Several colonies were Warren/Jonas sites. These surveys again found found in Lamar Valley, another was on the main the same three areas as Fullerton (southeast, stem of the Yellowstone River, and another was southwest, and northwest) densely populated by typically present north of Mammoth Hot Springs beavers. Beaver activity and sign was associated varying in location between Slide Lake and the with aquatics, willow (mostly), and aspen, but Beaver Ponds hiking trail with activity along the they did not provide locations for aspen. These Gardner River, but this colony was not present the areas were likely from the interior, not the Warren/ last two surveys. In 2009 a colony was located Jonas colonies. along Crystal Creek from the aerial survey, but later ground reconnaissance found that the bea- New Beaver Surveys vers did not overwinter at this site. None of the In 1996, we began complete park-wide aerial colonies anywhere in the park was found using surveys every other year completing 8 surveys aspen. A small number (< 5) of cottonwood were by 2009 (Figure 1). In 1996 we found only one observed cut into, but not toppled, in the Lamar beaver colony on the NR (Slide Lake near Mam- Valley, and several (< 5) small cottonwood trees moth), but after that, colonies began to slowly were observed downed along the Gardner River increase, especially along Slough Creek, and by north of Mammoth Hot Springs, otherwise all 2007 ten colonies were found on the NR (Table colony sites were associated with willow (Figure 2).

Beaver Abundance and Distribution in Yellowstone 283 The aerial counts found, similar to the Ful- in colonies from the late 1990s through 2009 lerton and Consolo Murphy surveys, three areas (Figure 1). Therefore, beaver populations on the of high colony density: the delta region along the NR appear to have been abundant in the early Yellowstone River extending to Thorofare, the 1900s, then the population declined through the Bechler region, and the northwest corner of the 1950s to the early 1990s and then increased from park. The delta region consistently supported the the late 1990s to 2009 (Table 2). It is notable that highest colony density in the wide willow plain that recolonization did not take place in the same areas, characterizes the Yellowstone River delta (Figure nor did it rely on aspen, which was the primary 1). Other regions of the interior supporting beaver woody vegetation used by beavers in the early colonies were Willow Park, Glenn Creek, Snake 1900s (Warren 1926); instead recovery relied on River, Shoshone Lake, Heart Lake, Lewis Lake willow. Most of the reported beavers in the1920s area (south), Harlequin Lake, Chipmunk Creek, were in the Tower Junction area (Warren 1926), Grouse Creek, Outlet Creek, and Hayden Valley whereas the recent recovery found most of the to name the most prominent and consistent colony beavers in Slough Creek and secondarily in the locations (Table 2). Most of the stream sites men- Lamar River (Table 2), with none in the Tower tioned did not have a beaver colony during each Junction area (except sporadic sign on Elk Creek survey occasion, suggesting movement or rotation with no permanent colony). The loss of aspen ap- among sites year–to–year or among surveys. All pears to be related to the beaver decline between of the colonies were located in willow stands the Warren and Jonas surveys, as emphasized (Figure 2), except the Harlequin Lake site where by Jonas (Warren 1926, Jonas 1955). The recent lodgepole pine was observed several years in the recovery was triggered by the late 1990s release food cache. of willow on the NR (Ripple and Beschta 2006, Beyer et al. 2007). It is unknown, but also unlikely, From 1996–2009, colony counts showed an whether the beaver recovery is as extensive or increasing trend (Table 2). In 1996, 49 colonies the population size as large as it was in the late were located in YNP, by 1999–2001 that number 1800s or early 1900s, or prior to the fur trapping had increased to 77 colonies park-wide, increasing of the mid-1800s (Schullery and Whittlesey 1992, again to 85 in 2003 (Table 2). Early ice-up cur- Wagner 2006) which raises the question of how tailed completion of the 2005 survey, but colony significant will be the ecosystem effects of the numbers had increased in 2007 to 127 colonies recent beaver recovery (Wolf et al. 2007). and were approximately stable between 2007 and 2009 with 116 colonies counted (Table 2). What caused willows and beavers to increase now after decades of low willow stature and few Tyers from 1986 to 1999 released 129 beavers beavers? Beaver abundance during Warren’s (1921 into drainages north of YNP. Ten (8%) of these and 1923; Warren 1926) surveys was based on beavers were implanted with radio transmitters, availability of aspen (and willow) created by a surge and 3 (30%) of this group likely dispersed into in aspen recruitment from 1870–90 (Romme et YNP, but the exact ultimate location is not known. al. 1995, Wagner 2006), but after beavers cut this aspen, it did not regrow, probably because of elk Discussion browsing (Larsen and Ripple 2003, Kauffman et al. More beaver colonies were found on the NR in 2010). Willow growth and stature paralleled those the early 1920s than in the 1950s and 1980s, al- of aspen (Wolf et al. 2007). Hence, the primary though the latter survey mostly recorded beaver woody vegetation that beavers relied on early in sign (not colonies), but was still suggestive of a the 20th century was gone, causing colony aban- smaller beaver population (Warren 1926, Arnold donment and a population decline. Competition 1931, Jonas 1955, Fullerton 1980). Evidence of with elk has been widely cited as a likely cause more sign was reported during the 1990s (Consolo for reduced woody vegetation (Jonas 1955, Kay Murphy and Hanson 1993, Consolo Murphy and 1990, Chadde and Kay 1991, Singer et al. 2000, Tatum 1994) and our surveys found an increase Wagner 2006, Ripple and Beschta 2006, Beyer

284 Smith and Tyers et al. 2007, Wolf et al. 2007), but other others if elk affect willow they are only the proximate suggest an interaction between elk and climate factor (YNP 1997). If true, then beaver popula- (YNP 1997, Johnston et al. 2011). tion fluctuations should be independent of elk, but Baker et al. (2005) specifically studied this rela- beavers fluctuated inversely with elk supporting tionship through rigorously controlled experiments the competitive hypothesis, but these correlative in Rocky Mountain National Park, concluding that data do not imply a cause and effect. Graumlich beaver cutting with intense elk browsing produced et al. (2003) found Yellowstone river flows in the th low stature willows (biomass and diameter) and 20 century exceeded those for the previous two “strongly suppressed the standing crop” (p. 110). centuries (therefore, and other than the 1930s, not They also referred to a system where low densi- a drying climate), and more importantly Wolf et al. ties of ungulates (both wild or domestic) caused (2007) found that, “climate is not limiting willow establishment in the study area” (p. 1582). They beavers and willows to flourish and described found that climate-driven hydrologic variables beaver–willow relationships (especially tall wil- explained only 11% of the variability in willow lows) as mutualistic. establishment and that nearly continuous willow In YNP, Bilyeu et al. (2008), tested these rela- establishment in the 20th century is not sensitive to tionships finding that loss of beavers and stream climatic variation. Further they state that substrate, incision “retarded willow height recovery” (nega- not water availability, limits seedling establish- tive feedback loop) emphasizing the importance of ment (Wolf et al. 2007). Of course climate and beaver induced hydrology and stated that, “very elk browsing could be interactive (Johnston et al. heavy elk browsing suppresses willows, and that 2011) and is an expanding area of study not yet reducing or eliminating browsing promotes willow fully understood. Other studies have also rejected recovery” (p. 89). Given this relationship, Wolf et the climate hypothesis (Chadde and Kay 1991, al. (2007) hypothesized that the loss of beavers Beschta 2005, Kauffman et al. 2010). altered the condition of the NR from a ‘beaver–wil- Besides changes in willow, the GNF beaver low’ state in the 1920s to an ‘elk–grassland’ state reintroduction from 1986 to 1999 (Smith and afterward. This elk–grassland state was present Tyers 2008) clearly hastened beaver occupation into the 1990s when some willows recovered of the NR of YNP, but it did not cause the beaver and some beavers returned, but whether a large recolonization. Beavers released north of the park scale landscape switch back to the previous state emigrated downstream into YNP speeding up re- (positive feedback loop) will take place is unclear occupation. Had GNF not reintroduced beavers, (Wolf et al. 2007). The above scenario appears movement of beavers in an upstream direction to be true for both the NR and the northwestern from the Yellowstone River would have occurred, region of the park, although tall willows are much just at a slower rate. Either way, recolonization more common in the northwest and more similar of beavers was dependent on recovery of willows to the beaver–willow state described by Wolf et their main building material and food source. Con- al. (2007). solo Murphy and Hanson’s (1993) observations An alternative hypothesis is that the scenario of beaver sign on the NR in the late 1980s and described above was due to a warming, drying early 1990s without extensive settlement suggests climate (especially during the 1930s), which dispersal of beavers into the NR of YNP from the is part of the long-term climatic cycle in YNP GNF upstream reintroduction prior to full willow (Persico and Meyer 2009). In responding to this recovery. Consolo Murphy and Hanson (1993) cycle, where fire was suppressed and conditions describe beaver movement as ephemeral and fre- were unfavorable to willow–aspen growth and quent and hardwood and willow communities as establishment, beavers consequently declined not abundant further indicating beaver habitat was (Houston 1982, Despain et al. 1986, YNP 1997). not yet suitable for the GNF releases to settle the Elk browsing (and wolf presence) is considered park. We also tracked beavers that had been radio of minor or little consequence to vegetation, and marked downstream into YNP, further confirming

Beaver Abundance and Distribution in Yellowstone 285 the source population for beaver settlement on NR population increases have already been at- the NR was from released GNF beavers (Tyers, tributed to willow recovery, but causes for beaver unpublished data). population increases in the interior are less clear, The population fluctuations of beavers in the particularly in the northwest (Figure 1, Table 2), interior of YNP is less clear because of less effort, but it is likely willow in the interior reflected well especially during the early surveys (Warren did not documented willow growth on the NR. It is pos- work in the interior; Figure 1). It appears beaver sible that below normal precipitation during the numbers have been stable here and subsisting on decade of the 2000s (Beyer et al. 2007, Wolf et willow (Figure 2, Fullerton 1980). Willow is less al. 2007) created drought conditions that might prone than aspen to depletion by beavers (Boyce have favored beaver colonization of higher gradi- 1974), and ungulate density is lower in the interior ent streams. Typically high water flows in spring especially in winter when browsing usually takes remove beaver across streams that were built place (Houston 1982, Wagner 2006), both of which the previous fall when the water was low. Without are factors that favor a stable food source and a high flows over an extended period of time in the less cyclical population. Despite variable survey spring, beavers may have been able to quickly re- intensity, beavers were found on every survey in build dams less damaged by reduced peak floods, the interior, and during the late 1990s and 2000s especially if the period of high flow was reduced. at high density in some willow stands (Table 2). It is known that abundant water improves beaver From 1996 on no beaver colonies in the interior habitat in low gradient areas (Baker and Hill were found utilizing aspen. It is unknown when 2003, Hay 2010), and although we acknowledge aspen became unavailable (Consolo Murphy and speculative, high flows may restrict settlement in Hanson 1993), as recent aerial surveys found mountainous areas because beavers cannot use aspen near beaver drainages but out of reach, and steeper gradient streams, thus periods of low flow observation of old cut aspen stumps suggested may allow beaver population expansion. Such a previous use. These observations were all made scenario may have been a contributing factor to along northwestern stream drainages and not in the a beaver population increase in the northwestern area of extensive willow at the Yellowstone River region of the park, but willow recovery was also delta, or in the Thorofare, Bechler, Snake, Willow a likely factor. Park regions which are also areas dominated by willow (Figure 2). Acknowledgements The recent (1996–2009) park-wide increase in We thank E. Albers, E. Almberg, and D. Guernsey beaver colonies was probably due to two factors: for help with database management and mapping 1) improved ability and efficiency of observers beaver colonies from both historical and current and 2) changing environmental conditions favor- surveys. M. Metz, K. Quimby and D. Stahler ing beaver expansion. The increase in numbers of assisted with aerial surveys. R. Stradley capably colonies found between 1996 and 1999 (Table 1) and safely handled all piloting duties on the suggest improvements in the ability of observers aerial surveys. We also appreciate presubmission to detect colonies, with probably a slight increase reviews from B. Baker and P. J. White. Funding in the beaver population. Between 1999 and 2003 and support were provided by the Yellowstone colony numbers were approximately stable, then Park Foundation and Yellowstone National Park. they increased significantly afterward (2005 was an Finally, the manuscript was improved due to the incomplete survey), suggesting an actual increase comments of two anonymous reviewers from rather than an improvement in observer/survey Northwest Science. efficiency as this had peaked by 1999.

286 Smith and Tyers Literature Cited River flow and teleconnections with Pacific basin climate variability during the past three centuries. Arnold, E. R. 1931. Beaver report: Tower Falls District– Climatic Change 59:245-262. Preliminary observations. Unpublished report on Hay, K. G. 1958. Beaver census methods in the Rocky file at Yellowstone National Park. Mountain Region. Journal of Wildlife Manage- Baker, B. W., and E. P. Hill. 2003. Beaver (Castor canaden- ment 22:395-402. sis). In G. A. Feldhamer, B. C. Thompson, and J. Hay, K. G. 2010. Succession of beaver ponds in Colorado A. Chapman (editors), Wild mammals of North 50 years after beaver removal. Journal of Wildlife America: Biology, Management, and Conservation, Management 74:1732-1736. The Johns Hopkins University Press, Baltimore and Hodgdon, H. E., and R. A. Lancia. 1983. Behavior of the London. Pp. 288-310. , Castor canadensis. Acta Baker, B. W., H. C. Ducharme, D. C. S. Mitchell, T. R. Zoologica Fennica 174:99-103. Stanley, and H. R. Peinetti. 2005. Interaction of Houston, D. B. 1982. The Northern Yellowstone Elk: beaver and elk herbivory reduces standing crop and Management. Macmillan, . of willow. Ecological Applications 15:110-118. Innis, H. A. 1930. The Fur Trade in . University Bangs, E. E., and S. H. Fritts. 1996. Reintroducing the of Press. Toronto. gray wolf to central Idaho and Yellowstone National Jenkins, S. H., and P. E. Busher. 1979. Castor canadensis. Park. Wildlife Society Bulletin 24:402-413. Mammalian 120:1-8. Beschta, R. L. 2005. Reduced cottonwood recruitment Johnston, D. B, D. J. Cooper, and N. T. Hobbs. 2011. following extirpation of wolves in Yellowstone’s Relationships between groundwater use, water northern range. Ecology 86:391-403. tables, and recovery of willow on Yellowstone’s Beyer, H. L., E. H. Merrill, N. Varley, and M. S. Boyce. northern range. Ecosphere 2:1-11. 2007. Willow on Yellowstone’s northern range: Jonas, R. J. 1955. A population and ecological study of the Evidence for a ? Ecological Ap- beaver (Castor canadensis) of Yellowstone National plications 17:1563-1571. Park. M.S. Thesis, University of Idaho, Moscow. Bilyeu, D. M., D. J. Cooper, and N. Thompson Hobbs. Kauffman, M. J., J. F. Brodie, and E. S. Jules. 2010. Are 2008. Water tables constrain height recovery of wolves saving Yellowstone’s aspen? A landscape- willow on Yellowstone’s northern range. Ecological level test of a behaviorally mediated trophic cas- Applications 18:80-92. cade. Ecology 91:2742-2755. Boyce, M. S. 1974. Beaver population ecology in inte- Kay, C. E. 1990. Yellowstone’s northern elk herd: A criti- rior Alaska. M.S. Thesis, University of Alaska, cal evaluation of the ‘natural regulation’ paradigm. Fairbanks. Ph.D. Dissertation, Utah State University, Logan. Chadde, S., and C. Kay. 1991. Tall-willow communities on Larsen, E. J., and W. J. Ripple. 2003. Aspen age structure in Yellowstone’s northern range: a test of the “natural the Northern Yellowstone Ecosystem, USA. Forest regulation” paradigm. In R. B. Keiter and M. S. Ecology and Management 179:469-482. Boyce (editors), The Greater Yellowstone Ecosys- Lemke, T. O., J.A . Mack, and D. B. Houston. 1998. Winter tem: Redefining America’s Wilderness Heritage. range expansion by the northern Yellowstone elk Yale University Press, New Haven, CT. Pp.231-262. herd. Intermountain Journal of Sciences 4:1-9. Consolo Murphy, S., and D. D. Hanson. 1993. Distribution National Research Council. 2002. Ecological Dynamics on of beaver in Yellowstone National Park. In R. S. Yellowstone’s Northern Range. National Academy Cook (editor), Ecological Issues on Reintroducing Press, Washington, D.C. Wolves into Yellowstone National Park. Scientific Nute, G. L. 1978. Caesars of the Wilderness. Minnesota Monograph, Denver, CO. Pp. 38-48. Historical Society Press. St. Paul. Consolo Murphy, S., and R. B. Tatum. 1995. Distribution of Persico, L., and G. Meyer. 2009. Holocene beaver - beaver in Yellowstone National Park, 1994. Unpub- ming, fluvial geomorphology, and climate in Yel- lished report on file in Yellowstone National Park. lowstone National Park, Wyoming. Quarternary Despain, D. 1990. Yellowstone Vegetation, Consequences Research 71:340-353. of Environment and History in a Natural Setting. Ripple, W. J., and E. J. Larsen. 2000. Historic aspen re- Roberts Rinehart, Boulder, CO. cruitment, elk, and wolves in northern Yellowstone Despain, D., D. Houston, M. Meagher, and P. Schullery. National Park, USA. Biological Conservation 1986. Wildlife in Transition: Man and Nature on 95:361-370. Yellowstone’s Northern Range. Roberts Rinehart, Ripple, W. J., and R. L. Beschta. 2004a. Wolves, elk, wil- Boulder, CO. lows, and trophic cascades in the Upper Gallatin Fullerton, S. 1980. Summary and impressions from beaver Range of southwestern Montana, USA. Forest survey (1979-1980). Unpublished report on file in Ecology and Management 230:161-181. Yellowstone National Park. Ripple, W. J., and R. L. Beschta. 2004b. Wolves and the Graumlich, L. J., M. J. Pisaric, L. A. Waggoner, J. S. ecology of fear: Can risk structure eco- Littell, and J. C. King. 2003. Upper Yellowstone systems? BioScience 54:755-766.

Beaver Abundance and Distribution in Yellowstone 287 Ripple, W. J., and R. L. Beschta. 2006. Linking wolves to Smith, D. W. 1997. Dispersal strategies and cooperative willows via risk-sensitive by ungulates in breeding in beavers. Ph.D. Dissertation, University the northern Yellowstone ecosystem. Forest Ecology of Nevada-Reno, Reno, NV. and Management 184:299-313. Smith, D. W., T. D. Drummer, K. M. Murphy, D. S. Guern- Ripple, W. J., and R. L. Beschta. 2007. Restoring Yellow- sey, and S. B. Evans. 2004. Winter prey selection stone’s aspen with wolves. Biological Conservation and estimation of wolf kill rates in Yellowstone 138:514-519. National Park, 1995-2000. Journal of Wildlife Romme, W. H., M. G. Turner, L. L. Wallace, and J. S. Management 68:153-166. Walker. 1995. Aspen, elk, and fire in northern Yel- Smith, D. W., and D. B. Tyers. 2008. The beavers of Yel- lowstone National Park. Ecology 76:2097-2106. lowstone. Yellowstone Science 16:4-15. Schullery, P., and L. Whittlesey. 1992. The documentary Swenson, J. E., S. J. Knapp, P. R. Martin, and T. C. Hinz. record of wolves and related wildlife species in 1983. Reliability of aerial cache surveys to monitor Yellowstone National Park area prior to 1882. In beaver population trends on prairie rivers in Mon- J. D. Varley and W. G. Brewster (editors), Wolves tana. Journal of Wildlife Management 47:697-703. for Yellowstone? A Report to the Wagner, F. H. 2006. Yellowstone’s Destabilized Ecosystem: Congress, volume 4, Research and Analysis. NPS, Elk Effects, Science, and Policy Conflict. Oxford Yellowstone National Park, Wyo. Pp. 1.3-1.174. University Press, New York. Shelton, P. C. 1966. Ecological studies of beavers in Isle Warren, E. R. 1926. A study of the beaver in the Yancey Royale National Park, Michigan. Ph.D. Thesis, region of Yellowstone National Park. Roosevelt Purdue University, Lafeyette, Indiana. Wildlife Annual 1:1-191. Singer, F. J., L. C. Zeigenfuss, and D. T. Barnett. 2000. Wolf, E. C., D. J. Cooper, and N. T. Hobbs. 2007. Hydro- Elk, beaver, and the persistence of willows: re- logic regime and herbivory stabilize an alternative sponse to Keigley (2000). Wildlife Society Bulletin state in Yellowstone National Park. Ecological 28:451-453. Applications 17:1572-1587. Smith, R. B., and L. J. Siegel. 2000. Windows into the Yellowstone National Park. 1997. Yellowstone’s Northern Earth: The Geologic Story of Yellowstone and Range: Complexity and Change in a Wildland Grand Teton National Parks. Oxford University Ecosystem. National Park Service, Mammoth Hot Press. New York. Springs, WY.

Received 24 June 2011 Accepted for publication 2 April 2012

288 Smith and Tyers